Nonlinear Wave Propagation Analysis using Experimental Data for the Estimation of Central Aortic Blood Pressure using Peripheral Circulatory Signals

University Of Maryland, College Park, College Park MD

Investigators

Abstract

Central aortic blood pressure significantly better predicts cardiovascular risk than commonly used brachial artery blood pressure. However, it is not widely used in clinical practice since it is very difficult to measure. This project concerns fundamental research on mathematical modeling to enable estimation of central aortic blood pressure from as few as two non-invasive peripheral circulatory waveforms. If successful, this project will advance personalized healthcare and medical device industry sectors. In particular, it will improve the way patients with cardiovascular disease are treated via its low-cost and non-invasive method that allows better assessment of cardiovascular health and prompt detection of its impairment. It may also stimulate the medical device industry and in turn the overall economy by stimulating the development of cost-effective, deployable medical devices for cardiovascular health monitoring, which will ultimately lead to improved therapy and treatment. This project will also have broad impact on education and knowledge dissemination, by bridging the gap between engineering, systems biology and medicine, attracting women and underrepresented students to multi-disciplinary research, and broadly disseminating the knowledge gained from this project through a dedicated website. The objective of this project is to test the hypothesis that central aortic blood pressure can be estimated from as few as two non-invasive peripheral circulatory waveforms via model-based system identification. To fulfill this objective, cardiovascular system is viewed as a nonlinear dynamic wave propagation system with unknown input. In this project, a theoretical study to develop system identification and input de-convolution methods for a class of nonlinear wave propagation systems will be conducted, and then the validity of the methods will be examined by applying them to the problem of estimating central aortic blood pressure waveform from two non-invasive peripheral circulatory waveforms. This project will contribute to advance dynamic systems and control by solving a system identification problem for a class of nonlinear dynamic systems. Since the class of systems considered belongs to cardiovascular health monitoring applications, it will advance systems physiology initiatives. Especially, this project will benefit the development of data-based modeling methods for physiologic systems and neuroscience, such as condition monitoring of cardiovascular hemodynamics and assessment of autonomic-cardiac regulation function. Ultimately, this project may open up new opportunities for technological advances in pervasive and personalized medicine, which can ultimately improve the quality of life of human beings.